Mustapha Mojahid

Learn More
In this paper we present our work in the field of accessibility to visual information in digital documents for the blind people. Our method consists in detecting the significant visual information in a picture/document, extracting their properties and rearrange them for the specified accessible output. Our approach relies on the efforts made in the field of(More)
This paper describes the performance of the VITIPI word completion system through a text input simulation. The aim of this simulation is to estimate the impact of the linguistic knowledge base size through two metrics: the Key-Stroke Ratio (KSR) and the KeyStroke Per Character (KPC). Our study shows that the performance of a word completion is depending of(More)
OATAO is an open access repository that collects the work of Toulouse researchers and makes it freely available over the web where possible. Abstract. This paper presents a framework which aims at describing text formatting , based on a model coming from the field of logic and linguistics, the Textual Architecture Model [23]. The goal is to improve(More)
SmartKey is a new virtual keyboard animation for handheld devices designed to make key selection easier for people with situational and motor impairments. The nearest key to cursor expands in four directions using the available space of nearby keys. Target expansion is accomplished with an occlusion factor of 0% and without any sideways motion. The two(More)
La compréhension d'un texte s'opère à travers les niveaux d'information visuelle, logique et discursive, et leurs relations d'interdépendance. La majorité des travaux ayant étudié ces relations a été menée dans le cadre de la génération de textes, où les propriétés visuelles sont inférées à partir des éléments logiques et discursifs. Les travaux pré-sentés(More)
We use extractive multi-document summarization techniques to perform complex question answering and formulate it as a reinforcement learning problem. Given a set of complex questions, a list of relevant documents per question, and the corresponding human generated summaries (i.e. answers to the questions) as training data, the reinforcement learning module(More)
READ (which stands for Documents Architecture REstitution, in French), is a software allowing augmented access of tagged documents content. It was originally developed for a project aiming at evaluating informational and functional equivalents to text visual formatting and was very recently tested in a user study (results unpublished yet). The design(More)